ZiWei Yuan
Chengdu City, Sichuan, China
Summary
🤩
RockstarZiwei Yuan is a Senior Software Engineer based in San Jose with a strong track record building web applications, cloud-native infrastructure and AI-powered services across fintech and retail at PayPal and Walmart Global Tech. With a background in data informatics and geographic information systems, Ziwei combines rigorous backend skills in Java, Go and Python with practical experience in databases, NoSQL and Kubernetes to deliver scalable systems and automation. He has driven measurable impact—accelerating product generation by 82% with a Gen-AI microservice at Walmart and improving processing efficiency and latency in PayPal’s private cloud platform by leveraging event-driven design, Redis and Go profiling. An active open-source contributor, he has focused on back-end optimizations for LLM inference in the ktransformers project, including MOE improvements and chat-history fixes that boost model performance. Known for translating research-grade ideas into production-ready services, Ziwei blends systems engineering discipline with a curiosity for AI tooling and performance tuning.
5 years of coding experience
Github Skills (8)
transformers10
pytorch10
python10
back-end-development10
optimization9
c-language8
machine-learning8
cprogramming-language8
Programming languages (7)
Github contributions (5)
kvcache-ai/ktransformers
Sep 2024 - Apr 2025
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
Role in this project:
Back-end Developer
Back-end DeveloperContributions:8 releases, 53 PRs, 57 pushes in 6 months
Contributions summary:ZiWei primarily focused on back-end improvements and bug fixes within the `ktransformers` repository. Their commits addressed issues in the transformers interface, fixing chat history inaccuracies and refining message handling. Furthermore, they implemented support for older model configurations, and updated the local chat functionality. The user also made significant changes to the MOE (Mixture of Experts) code, indicating a focus on optimizing model performance.
KMSorSMS/KMSorSMS
Jan 2023 - Jan 2023
Contributions:28 commits, 25 pushes, 1 branch in 2 days
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.